Spatial and Temporal Variation of in-situ Stress in and around Active Fault zones in Central Japan

In the "Active Fault Zone Drilling Project in Japan," we have compared the relationship between the stress concentration state and the heterogeneous strength of an earthquake fault zone in different conditions. The Nojima fault which appeared on the surface by the 1995 Great Kobe earthquake (M=7.2) and the Neodani fault which appeared by the 1891 Nobi earthquake (M=8.0), have been drilled through their fault fracture zones. A similar experiment conducted on and research of the Atera fault, of which some parts have seemed to be dislocated by the 1586 Tensyo earthquake (M=7.9). We can use a deep borehole as a reliable tool to understand overall fault structure and composed materials directly. Additionally, the stress states in and around the fault fractured zones were obtained from in-situ stress measurements by the hydraulic fracturing method. Important phenomena such as rapid stress drop in the fault fracture zones were observed in the Neodani well (1300 m deep) and the Nojima well (1800 m) of the fault zone drillings, as well as in the Ashio well (2,000 m) in the focal area. In the Atera fault project, we have conducted integrated investigations by surface geophysical survey and drilling around the Atera fault. Four boreholes (400 m to 600 m deep) were located on a line crossing the fracture zone of the Atera fault. We noted that the stress magnitude decreases in the area closer to the center of the fracture zone. Furthermore the orientation of the maximum horizontal compressive stress was almost reverse of the fault moving direction. These results support the idea that the differential stress is extremely small at narrow zones adjoining fracture zones. We also noted that the frictional strength of the crust adjacent to the faults is high and the level of shear stress in the crust adjacent to the faults is principally controlled by the frictional strength of rock. We argue that the stress state observed in these sites exists only if the faults are quite "weak." As a temporal variation of stresses, crustal stress was recorded from 1978 to before the Kobe earthquake in and around the area where the earthquake occurred. By examining this data, the change in tectonic stress gradually increased prior to the earthquake. After the earthquake, the same boreholes were once again used to obtain new data. From these measurements, we were able to determine that there was a definite drop in the crustal stress in the area and that there was a change in the direction of the principal stresses. The continual measuring is essential to estimate the absolute stress magnitude that initiate earthquakes and control their propagation.